UNI-T UT131C serial data output.

[Refrigerator]

When changing the 10A fuse in my UT131C i noticed these big pads clearly labeled TX and RX.
There are four pads connected in pairs for TX and RX.
Probing the TX pad revealed that the meter is spitting out some serial data - neat!
So i connected my logic analyzer and found out that it's transmitting 10 bytes of data every 500ms at 800 2400 baud.
Didn't take long to notice that the 8-th byte contains the value measured in plain binary.
For example in the 20V range 1.09V will translate to a 109 in binary on the 8-th byte.
So it seems that unlike some other meters this doesn't just dump raw LCD data out the serial port.
Haven't yet delved any deeper into this, but there's sure to be range data and lots of other stuff in the 10 bytes of data that this thing spits out.
Also there's an RX pad on the meter as well, which i assume would be for calibrating?

Haven't found anyone else talking about this, am i the first one to find this out?
Personally i see some use for this for data acquisition and logging.
But the meter has auto power off that i don't know how to disable, perhaps if i tickled the RX port with some null data from time to time the meter wouldn't turn off?

EDIT: if you hold down the HOLD/SEL button when turning the meter on the auto power off disables! 

[Refrigerator]

More findings:
Bytes 7 and 8 form a signed 16-bit integer, which represents the measured value.

[Refrigerator]

More findings:
The value transmitted through serial can exceed the max display value of the LCD.
Example:
In 2000mV range my meter shows OL, but serial says 2137mV.
But if i switch the range to 20V it shows 2.16V, meaning that the serial data is most likely correct.

[Refrigerator]

Even more findings:
So i had an idea, if serial data is able to display numbers way of out the range of the LCD then does that mean i can use the entire range of the 16-bit signed integer?
The answer is no, but it's still pretty far out of the LCD range. How far out you ask?
The maximum value i was able to get in the 200mV range was 0x2056 (that's hex) or 827.8mV !!!!!! 
That's four times more than the range of the LCD 
And yes, i measured it and it seems accurate. 
Measuring that far out of range might lead to inaccuracy but does that also mean that this meter uses a 6000 count chipset?
In a 20€ meter? No way.
But that's great news for me because i've often found the 2000 counts on the meter to be a limiting factor, especially since it doesn't have auto range.

[Refrigerator]

So i went ham on all the ranges and copied all the byte values into an excel spreadsheet.
The table is not complete but gives enough info for now imo.
Bytes 9 and 10 seem to change depending on the measurement, i assume they're some sort of status flags.
Byte 4 seems to be the range.
Funny thing i noticed is that in ^oC range serial outputs ^oC but in ^oF it still outputs ^oC.
Also serial temperature data gives 0.1^oC resolution while the LCD gives only 1^oC resolution.
So measuring temperature through serial gives better resolution.

PS: i wrote byte values down in decimal in hopes to see if there's a pattern to them changing.

[Refrigerator]

Enabling hold or backlight does not seem to have any effect on the output data.

[ledtester]

In a lot of cases byte 10 is one more than byte 4.

Update: It appears that byte 10 is just the sum of bytes 3 through 8 -- i.e. like a checksum. In fact, bytes 9 and 10 might form a 16-bit checksum.

[Kean]

It is likely obvious to most, but connecting to these pads means lack of any isolation between the circuit & voltages being measured and serial interface.

I'd strongly suggest adding either an opto isolator, or better add an IR LED transmitter via a hole in the DMM case and close it back up.  Then build an IR receiver that can mate with the DMM case for reliable logging.

[Refrigerator]

It is likely obvious to most, but connecting to these pads means lack of any isolation between the circuit & voltages being measured and serial interface.

I'd strongly suggest adding either an opto isolator, or better add an IR LED transmitter via a hole in the DMM case and close it back up.  Then build an IR receiver that can mate with the DMM case for reliable logging.

That comes later.
For now i just need to find out what's going on the serial port.
Also there seems to be no need to use the RX port because the DMM resets as soon as anything happens there and there's no way to find out what can and can't be done through it.
After i'm done i'll put an IR LED in the back for comms just because it's more convenient than pin headers.